@inproceedings{bib_Dete_2023, AUTHOR = {Arti Yardi, Vamshi Krishna Kancharla, Amrita Mishra}, TITLE = {Detecting Linear Block Codes via Deep Learning}, BOOKTITLE = {Wireless Communications and Networking Conference}. YEAR = {2023}}

In the channel code detection problem, given a sequence of noise-affected codewords generated by an unknown code, the aim is to identify the correct channel code from the given set of potential codes. This problem has many applications in military spectrum surveillance and in cognitive radios. In this paper, we consider the situation when the set of potential channel codes consists of two linear block codes and propose a deep learning based classification approach for the corresponding code detection problem. In this work, we propose data processing strategies suitable for linear block codes that reduce the amount of data required to train the deep neural network classifier. This however comes at the cost of having a lower probability of detection. For the proposed data processing strategies, we analytically obtained the optimal probability of correct detection that one can hope to achieve using any neural …

@inproceedings{bib_Thir_2022, AUTHOR = {Arti Yardi}, TITLE = {Third-party cyclic code reconstruction over binary erasure channel}, BOOKTITLE = {International Journal of Information and Coding Theory}. YEAR = {2022}}

We consider a setup where Alice is transmitting channel coded messages to Bob and intruder Willie is snooping over this communication. Willie does not know the channel code that Alice is using and wishes to identify it in order to decode the intercepted messages. This problem is termed as blind reconstruction of channel codes and has possible applications in cognitive radios and military. In this work, we consider the situation when Willie knows that Alice's channel code belongs to the family of binary cyclic codes and the underlying communication channel is the binary erasure channel. We present an algorithm to identify the parameters of the unknown channel code corresponding to the given data. The key step in our algorithm consists of distinguishing between the two situations when all of the assumed parameters are correct and when either of the parameter is incorrect. As part of analysis, we present a lower bound on probability of correctly distinguishing between these two situations.

@inproceedings{bib_EBP-_2022, AUTHOR = {Arti Yardi, Tarik Benaddi, Charly Poulliat, Iryna Andriyanova}, TITLE = {EBP-GEXIT Charts for M-ary AWGN Channelfor Generalized LDPC and Turbo Codes}, BOOKTITLE = {EEE Transactions on Communications}. YEAR = {2022}}

This work proposes a tractable estimation of the maximum a posteriori (MAP) threshold of various families of sparse-graph code ensembles, by using an approximation for the extended belief propagation generalized extrinsic information transfer (EBP-GEXIT) function, first proposed by M ́easson et al. We consider the transmission over non-binary complex-input additive white Gaussian noise channel and extend the existing results to obtain an expression for the GEXIT function. We estimate the MAP threshold by applying the Maxwell construction to the obtained approximate EBP-GEXIT charts for various families of low-density parity-check (LDPC), generalized LDPC, doubly generalized LDPC, and serially concatenated turbo codes (SC-TC). When codewords of SC-TC are modulated using Gray mapping, we also explore where the spatially-coupled belief propagation (BP) threshold is located with respect to the previously computed MAP threshold. Numerical results indicate that the BP threshold of the spatially-coupled SC-TC does saturate to the MAP threshold obtained via EBP-GEXIT chart.